We propose to build upon our initial findings of factors that influence risk of nephrolithiasis, which occurs in 12 percent of the US population and causes tremendous pain and suffering. Using a prospective study design, the primary objective is to identify and evaluate dietary and urinary risk factors for recurrent nephrolithiasis in female and male participants of three large prospective cohort studies: the Nurses' Health Study I (121,700 women), the Nurses' Health Study II (116,671 women), and the Health Professionals Follow-up Study (51,529 men). Detailed dietary data are collected and updated in all three cohorts, and 24-hour urine samples were previously collected from a subset of subjects with incident kidney stones from all three cohorts. During the proposed 5-year study period, we will identify and confirm incident and recurrent kidney stones by medical record review. Further, we will collect two 24-hour urine samples from newly confirmed incident cases and randomly selected controls to measure 24-hour urine excretion of calcium, oxalate and other relevant factors. We will test the hypotheses that dietary and urinary factors proposed to be associated with incident stone formation are associated with recurrent stone formation. We also will explore interactions between dietary and urinary factors, especially calcium intake and urinary calcium excretion and risk of stone recurrence. Furthermore, we will examine if recurrence rates differ by gender and, if so, whether these differences can be explained by differences in dietary or urinary factors. Basing these analyses in these three existing cohorts has several major strengths. A substantial number of incident cases (n = 4953) and recurrent cases (n = 1114) have already been identified. The completeness of follow-up and the vast amount of information already collected make this a rich resource, particularly to study interactions between diet and urinary factors. Several important and intriguing results regarding kidney stone formation have already emerged from the cohorts. The study design and use of existing data provide an efficient means to increase the scientific value of these cohorts and to test several hypotheses of public health importance.